Real-Time Implementation of Squeezeseg V3 Semantic Segmentation Using A Vck190 Fpga Board

In the last few years, the automotive industry has relied heavily on deep learning applications for perception solutions. With data-heavy sensors, such as LiDAR, becoming a standard, the task of developing low-power and real-time applications has become increasingly more challenging. To obtain the maximum computational efficiency, no longer can one focus only on the software aspect of such applications, while disregarding the underlying hardware. This paper presents a hardware-software co-design approach used to im-plement an inferencing application using the SqueezeSegV3, an efficient convolutional neural network for 3D LiDAR point-cloud segmentation, on the Versal ACAP VCK190 FPGA. Using the complete 360° Semantic-KITTI dataset point-clouds, a real-time framerate of 11 Hz is achieved with a peak power consumption of 78 Watts and minimal accuracy loss. A smaller ver-sion of the same model is also deployed achieving a framerate and peak power consumption of approximately 19 Hz and 76 Watts respectively.